New #EMNLP2018 paper "Rapid Adaptation of NMT to New Languages" w/ Junjie Hu! Exciting results: training models on 50+ languages not only allows for good X-lingual adaptation, you can also get BLEU up to 15.5 with NO training data in the source language 😲 https://t.co/NOpwfM97cV pic.twitter.com/nIlBWJgHxk
— Graham Neubig (@gneubig) August 10, 2018
As a data scientist or NLP researcher or anything that vaguely works with large scale data, tell me you don't look at this tweet and then immediately start salivating ^_^https://t.co/iLQ6JR0PEX
— Smerity (@Smerity) August 8, 2018
you want to replace unlimited history with size-k history because you think it is a good enough approximation? by all means please do. but don't expect to magically capture also the long-range stuff. it doesn't work this way. it cannot work this way.
— (((ل()(ل() 'yoav)))) (@yoavgo) August 8, 2018
Also, many people seem to hold both of the following beliefs at the same time:
— (((ل()(ل() 'yoav)))) (@yoavgo) August 8, 2018
- ha cool we can do language models with feed-forward nets instead of RNNs!
- if we do LM well we will model all of language and achieve AGI!
It doesn't work this way. These are conflicting.
I made a short summary and a map to explore last week's fascinating twitter megathread on Semantics and Meaning in NLP.
— Thomas Wolf (@Thom_Wolf) August 7, 2018
You can find it here https://t.co/A9tlpc90wM
(To the participants: please, don't hesitate to tell me if you have remarks on how I report your arguments) pic.twitter.com/bJQzkBkGnX
"Large Scale Language Modeling: Converging on 40GB of Text in Four Hours," Puri et al.: https://t.co/EKMpNxdPSw
— Miles Brundage (@Miles_Brundage) August 7, 2018
New NLP News: Future of CV and NLP, Maths (Fields Medal, Maths for ML), AutoML Natural Language, Google Assistant, Cutting through the hype, NALUs https://t.co/dopMqE6LPV (via @revue)
— Sebastian Ruder (@seb_ruder) August 6, 2018
Also, play around with the data yourself: https://t.co/CHeRHx0Env
— Micah Cohen (@micahcohen) July 31, 2018
I'd love to publish cool things people do with it.
You can check out my slides from #jsm2018 today on text mining and modeling here: https://t.co/QndAIUWjbjhttps://t.co/qSYlbHcDlJ
— Julia Silge (@juliasilge) July 30, 2018
Two themes were most prominent for me at #ACL2018:
— Sebastian Ruder (@seb_ruder) July 26, 2018
1) Understanding representations.
2) Evaluating models in more challenging settings.
See below for the full summary. https://t.co/JxCDONw3qk
Just joined #NIPS2018 Conversational Intelligence challenge
— Thomas Wolf (@Thom_Wolf) July 25, 2018
… straight to the top of the leaderboard 🚀
If you're curious, you can chat with the 🤖 @ https://t.co/tB6EGE4CuR
Our models mix VAE, attention, MTL, IR & semantic tech. You know & like this stuff? Shoot me a DM! pic.twitter.com/Dn2P5Ifnm1
https://t.co/ikPqrzROnj website is up!
— Richard (@RichardSocher) July 25, 2018
Slides motivating true multitask learning in AI and NLP from a recent talk: pic.twitter.com/ChJTMvDkcC
